Project Abstract Background: Angiosarcoma is a rare and aggressive cancer for which new treatments are urgently needed. Efforts to implement precision medicine in this cancer are hindered by its rarity, disease heterogeneity, the time needed to collect informative clinical data, and the emergence of chemotherapeutic resistance. Liquid biopsy is a promising new technique allowing the sequencing of tumor DNA circulating in the plasma as a proxy for tumor biopsy. This noninvasive technique has great potential to improve patient outcomes by allowing for longitudinal monitoring. However, in order for this technique to be applied in the clinic, key questions regarding its optimal use must be answered. I propose to develop liquid biopsy as a clinical monitoring strategy in angiosarcoma using the pet dog as a model. ?Dogs commonly develop hemangiosarcoma, which is highly similar to angiosarcoma at the clinical, histopathological, and genomic level. The dog model thus offers the opportunity to implement rapid clinical trials for a rare cancer in a large patient population?. Objective/Hypothesis: ?I propose to advance the use of liquid biopsy as a viable clinical strategy to monitor treatment response, and to identify relapse earlier, by harnessing the strengths of the canine angiosarcoma model. Furthermore, I will address a major cause of patient mortality by characterizing mechanisms of doxorubicin resistance. Specific Aims: ?(Aim 1) I will validate canine hemangiosarcoma as a model for human angiosarcoma at the genomic level through detailed characterization of somatic mutations in whole-genome and whole-exome sequencing. (Aim 2) I will optimize protocols for liquid biopsy in a cohort of canine hemangiosarcoma patients in order to maximize the diagnostic quality of the samples, and implement and refine techniques for tracking minimal residual disease (MRD). (Aim 3) I will identify mutations associated with doxorubicin resistance. Resistance leading to disease progression is a major cause of patient mortality, and I will determine whether these mechanisms can be targeted using existing therapeutics. ?Study Design: Thirty dogs with splenic hemangiosarcoma will be enrolled prospectively at diagnosis. Tumor and normal samples will be collected at surgery, and longitudinal blood samples drawn at diagnosis, surgery, and at each doxorubicin treatment until relapse. (Aim 1) I will perform tumor/normal whole genome sequencing of a subset of cases, which will allow for characterization of regulatory and structural variants, and will compare these results to the whole exome data currently being generated by The Angiosarcoma Project. (Aim 2) I will develop and refine methods for sampling cfDNA for longitudinal monitoring of MRD, through testing of different phlebotomy protocols, and designing patient-specific hybrid capture probes. (Aim 3) ?I will identify mutations associated with doxorubicin resistance, by analyzing somatic mutations in the same patient at diagnosis and again at relapse.